基于MEEMD加窗改进方法与GRNN组合的电力负荷预测  

Power Load Forecasting Based on MEEMD Windowing Modification Method and GRNN Combination

作　　者：高晗 段艳涛 毕贵红[1] GAO Han;DUAN Yan-tao;BI Gui-hong(School of Electric Power Engineering,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学电力工程学院

出　　处：《软件导刊》2019年第12期44-50,共7页Software Guide

摘　　要：针对电力负荷序列不稳定性、随机性等特点引起的电力负荷预测精度下降等问题,提出MEEMD加窗改进方法和GRNN组合的短期电力负荷预测模型。利用GRNN神经网络延拓方法对原始信号两端数据进行延拓,用余弦窗函数对延拓数据加窗处理后再进行MEEMD分解,用神经网络对各分量趋势进行预测,叠加各分量的预测结果得到负荷序列的最终预测结果。实验结果表明,MEEMD加窗改进分解预测的平均绝对误差、平均绝对值百分比误差和均方根误差分别为73.9268、0.8180%和82.9301。基于MEEMD加窗改进方法和GRNN组合的电力负荷预测不仅能抑制端点效应,而且能解决模态混叠和伪分解问题,提高了短期电力负荷的预测精度。The forecasting accuracy of power load is influenced by the instability and randomness of power load sequence.To solve this problem,a short term load forecasting model based on MEEMD windowing modification and GRNN combination is proposed.First,the original signal data were extended at both ends by using the method of GRNN neural network extension.Then,the extended data were processed by Cosine window function.Finally,the windowed signal were decomposed by MEEMD method.The neural network was then used to predict the trend of each component,the ultimate forecasting results can be obtained by the superposition the forecasting results of each component.The experimental results show that the mean absolute error,mean absolute percentage error and root mean square error of MEEMD windowing modification decomposition prediction are 73.9268,0.8180%and 82.9301,respectively.Power Load Forecasting Based on MEEMD Windowing modification method and GRNN Combination can not only suppress the end effect,but also solve the problem of mode mixing and pseudo decomposition,and improve the forecasting accuracy in short-term power load fore?casting.

关 键 词：电力负荷预测 GRNN 端点效应 模态混叠 余弦窗函数 

分 类 号：TP306[自动化与计算机技术—计算机系统结构]